Cycle Analyst V3 preview and first beta release

[teklektik]

OK, implemented AutoTorqPAS. It basically worked but still had the same dropout problem .
...
I flashed back in Prelim3 firmware and did a basic RPM PAS config. The RPM PAS worked perfectly without any dropout issues at all. Next I tried AutoTorquePAS with prelim3 firmware. This worked really well with no dropouts and operated just like AutoPAS on the latest version. Still needs some fine tuning but is looking good at the moment.

Interestingly, with the latest firmware, I couldn't get speedPAS to work at all. I could see the cadence bars going up and down but the motor would never activate. At the time, I didn't think too much about it as I figured i had missed a setting and was concentrating on AutoPAS anyway which did work.

Anyway, I have run out of options with Prelim5 and magnetic ring PAS. for now I will stick with B23 on the bike using a magnetic ring PAS and Prelim5 on my other bike with Torque PAS.

Wow! Busy afternoon.
Thanks for all the tests - particularly the rollback tests to v3.0P3 which seem pretty definitive. Good stuff.

Too bad the AutoTorqPAS trick didn't get around the issue, but it's good to see that it had a successful road trip . You may have noticed that this is just a special case of the 'Tips and Tricks' gimmick in the Guide where the PAS Assist pot is attached to the Trq input so you can use a 3-pos switch at the same time for throttle control. Here we just left off the pot and rely on the pull-up.

Last thoughts: The only related thing that came up in testing and hasn't been posted was an issue with ThrI->MinInput being set too close to the actual Throttle minimum (dead zone too small; offset less than 0.1v). This is similar to the issue posted here a bit ago about an auto-cruise problem. With the offset too small, a little noise would cause the CA to think the throttle was applied briefly, knocking you out of auto-cruise. I had a similar problem in PAS testing where I couldn't get a PAS mode to start because the CA thought the throttle was engaged. Oddly, everything worked fine in the previous release but a subtle change in v3.0P5 caused this 'throttle detection' issue. I really doubt that this is in play in your case, but scraping the bottom of the barrel for ideas, it's the last config issue I can think of that sounds even remotely applicable.

 

[Kepler]

Wow! Busy afternoon.
Thanks for all the tests - particularly the rollback tests to v3.0P3 which seem pretty definitive. Good stuff.

Too bad the AutoTorqPAS trick didn't get around the issue, but it's good to see that it had a successful road trip . You may have noticed that this is just a special case of the 'Tips and Tricks' gimmick in the Guide where the PAS Assist pot is attached to the Trq input so you can use a 3-pos switch at the same time for throttle control. Here we just left off the pot and rely on the pull-up.

Yes I had read that section quite carefully and thought it was a good idea. I like the modified version relying on the pull-up even better though as it achieved what i was looking for without the need to add or modify my existing wiring. 8)

Last thoughts: The only related thing that came up in testing and hasn't been posted was an issue with ThrI->MinInput being set too close to the actual Throttle minimum (dead zone too small; offset less than 0.1v). This is similar to the issue posted here a bit ago about an auto-cruise problem. With the offset too small, a little noise would cause the CA to think the throttle was applied briefly, knocking you out of auto-cruise. I had a similar problem in PAS testing where I couldn't get a PAS mode to start because the CA thought the throttle was engaged. Oddly, everything worked fine in the previous release but a subtle change in v3.0P5 caused this 'throttle detection' issue. I really doubt that this is in play in your case, but scraping the bottom of the barrel for ideas, it's the last config issue I can think of that sounds even remotely applicable.

Tried this already. I had set my throttle input / output really tight for minimum throttle dead band. reading your solution on the auto-cruise problem, I thought it may be similar also. So I loosened up the MinInput from 1.30V to 1.20V. I could then feel the slight increase in dead-end on the throttle. I even activated the Auto-cruise and did a heap of testing with that to see if I would see the same drop out issue. In auto-cruise I experience no dropouts at all.

 

[lizardboy]

Hi all;

just had a new problem crop up that has me a bit stumped. When I was out riding just now the bike started cutting out randomly just for a second or two. I use current control throttle on my setup and have been happy with it. Today when I was riding, the bike would cut out randomly for a second every now and then. I changed to the screen that shows the throttle in and out voltages. I observed that when the cut out happens the throttle out voltage drops by more than a volt causing the drop out. The input voltage stays normal. The "S" also changes to a capital indicating that speed limiting is active. My speed limit is set to 199 km/h so I'm pretty sure I'm not exceeding that. I did a bit more checking and it seems that when looking at the first screen I can see the kph label flashes and the speedo indication changes to zero when the cut out happens. This indicates to me that the speed sensor is broken somehow or going intermittent this I can probably deal with not sure why a sealed sensor would crap out but I can accept it. What I don't understand is why a speed sensor is causing the throttle signal to cut out. It should be irrelevant to the throttle output in current mode right? I also tried speed and pass-thru mode which changed nothing. It's an old gripe of mine but why can't pass through mode actually act as a pass through? I'd like to see a hardware bypass or at least have pass through ignore things like limits. I've checked over all the wiring and there is nothing wrong I can see. All I know for sure is that the output voltage goes down for no apparent reason. I could use some direction and some explaining to get me on the right track. It's worked pretty flawless since the software bug that was causing my initial cut out problem got fixed

 

[Kepler]

Timing issues fixed that could cause occasional "T_Err" under some special circumstances

Problem with rolling back one version is that the above error is back on the Fighter Shutting down temp sense stops the error we all all now know but temp sense is an important feature to me.

 

[teklektik]

When I was out riding just now the bike started cutting out randomly just for a second or two.
...The "S" also changes to a capital indicating that speed limiting is active.
...My speed limit is set to 199 km/h so I'm pretty sure I'm not exceeding that.
...when looking at the first screen I can see the kph label flashes
...This indicates to me that the speed sensor is broken somehow

The cap 'S' means speed limiting is in play - this is what is cutting your throttle. The flashing 'kph' indicates that you have exceeeded MaxSpeed.

So - it's a pretty safe bet that your speedo sensor reed switch has developed contact bounce. When the magnet goes by, it gives out a flock of tiny pulses that make the CA think you are going more than 199kph. There is no good fix of which I am aware except replacement. You can order another pickup from Grin or use another from a cheapie bike computer if you have an old one on the shelf.

If the symptom persists (it probably will...), an interim fix until replacement is to disconnect the sensor, remove the spoke magnet, or tie-wrap the sensor further up the fork where the magnet won't trigger it.

The best alternative if you are using the CA-DP cable to the controller is to pull the sensor and switch to using the controller hall signal. The Guide explains CA mod - very simple.

 

[lizardboy]

Thanks for the advice it's a geared motor so that won't work long term. Figures I just sold my old cycle analyst last week and when I got the new CA I just resused the old sensor to save taking the wiring harness apart that seems like a dumb idea now....Another good fix would be to have a setting of 199kph actually disable the speed limiter so no one gets stranded. This is what I'm saying about having a true pass through setting that ignores the limits. Kind of dumb to get stranded because of a speed sensor when the intent of my settings was to disable it anyway as I don't need it. I think I've had just about every failure you can have at one point or another and I could really see how a few limp home settings would make the difference in getting home or walking. Failures of the inputs to the CA should be able to be overcome on the road. Maybe even a failed throttle could have some sort of on/off use of the screen buttons to get you home there's of course the speed sensor and maybe if the whole CA fails due to power supply or the like it would be nice if things defaulted to a pass through of the throttle signal. This maybe more doable simple even on my bike as I power the throttle from the controller. Just ideas no one has to listen
thanks for the help.

 

[hjns]

A temporary solution is just to remove the magnet or move it anywhere else, so that the speed sensor is not triggered anymore. Of course you will loose the speed readout, but at least you dont have the interupts anymore.

 

[teklektik]

This maybe more doable simple even on my bike as I power the throttle from the controller.

Exactly right, and a single throttle bypass addresses all CA failure modes - it's not even necessary to install/mount a switch, just add another JST-3 connector. I have a pass-through throttle connector under then dash to drive the controllers directly. The fallback is move the throttle connector to the by-pass connector. Done. I've never had to use it, but it's tucked away if the need arises.

 

[hjns]

This maybe more doable simple even on my bike as I power the throttle from the controller.

Exactly right, and a single throttle bypass addresses all CA failure modes - it's not even necessary to install/mount a switch, just add another JST-3 connector. I have a pass-through throttle connector under then dash to drive the controllers directly. The fallback is move the throttle connector to the by-pass connector. Done. I've never had to use it, but it's tucked away if the need arises.

Yes that will work. However, it will be a big scare on a high-powered vehicle when you are used to a CA power-throttle and then have to manage the speed throttle. But it will get you home, and that is what counts!

 

[Merlin]

Hey, iam Back. Just want to let you know that your Post
helps me (again) alot. On my 3 Presets i can go now on the legal (1) Preset with Throttle (speed) and Allday (2) Preset with PassThru.
(on 2nd theres no Speed Limit)
The Problem between Speed and Path-Thru is that on "Speed" i cant precisely control the throttle.
giving a little (20%) the bike gives everything it has....maybe clear to other ones. but not to me really clear. So i needed to use Pass thru on Preset 2.
The Tricky part was for me to set up both with only one PS/DS/IntS Gain.
I spend 100km in start/stop n go change settings and trying. GodD*mn It by butt hurts, you cant set up the bike without real load :x
The Main Reason for Dropouts was the "Dead Zone" as described...after that it was alot better.
From Beta23 to PreLim5 there MUST be something changed with PAS/Throttle...i had the same experience that it works alot better with B23.
My First Preset (Legal....25kph 250w) satisfy my claims to 95% ...i cant get it perfect because i wont change the DS Gain higher than 30. So the Speedlimit works a bit laggy.
Good Enough for a testride....better than every Wallmart Bike....good enough for a testing cop
It is only the Fallback Preset and iam using 2 & 3.
I think with Seperate Speed Settings (DS/PS/IntSGain) for each Preset you can setup it perfect. Maybe iam to fussy in this point. :wink:

If we need Program-Space for seperate DS-PS-IntS-Gain i dont need other Throttle Modes than Auto PAS (who has nothing to do with torque)
(... someone who is using Throt-PAS say a loud "here" )
I dont know how much Preset Names are using. my suggestion is 5-6 are enough.
Another Idea to get more Programm-Space....Battery Chemistry.....
Who is using NIMH accus?!
whats the difference between Lipo and LiFe ?

I was afraid you had another preset, but thought it was important to do the DSGain setting to identify the problem before making things more complicated. Now we focus on getting your SGain settings fixed.

You are having the same problem that Sebbo had a while back. The answer is to properly adjust IntSGain, PSGain, and DSGain instead of simply setting them to deactivate all speed control. If you can get these properly set, the settings will work for all presets.

Please have a look at this post. It describes a setup procedure for speed settings for powerful bikes like yours. My bike is 3.3kW with gear motors and so has huge torque off the line - it accelerates rapidly and has lots of overshoot. This procedure was developed on that bike and seemed to work fine, but hasn't been tested on other bikes so I can make guarantees (the lack of testing is why it does not yet appear in the Guide).

Please give this a try and see how it works - do the setup without PAS. It's experimental so No Promises - but I would be interested in your feedback.

 

[teklektik]

The Problem between Speed and Path-Thru is that on "Speed" i cant precisely control the throttle.
giving a little (20%) the bike gives everything it has....maybe clear to other ones. but not to me really clear.
So i needed to use Pass thru on Preset 2.

Yep - Speed throttle behaves in a way that seems very unnatural. The throttle is setting the speed between 0-100% of the configured MaxSpeed and the CA tries to accomplish that by applying full power until it gets close. It's pretty startling on a powerful bike.

But - it's not necessary to use Speed Throttle - the same gain settings work for simple speed limiting in other throttle modes, so once you are tuned up, all will be good. You might try switching to Power or Current Throttles in your presets to make the throttle more manageable - while your existing speed gain tuning should continue work okay to do the speed limiting.

From Beta23 to PreLim5 there MUST be something changed with PAS/Throttle...i had the same experience that it works alot better with B23.

Yep - looks like P6 may need a bit more attention than originally foreseen. Justin is traveling and will need to get general business matters in order when he returns, so it's great that you were able to get a usable setup in place for the interim.

Anyhow - happy to see you are up and running. Thanks for the update!

 

[teklektik]

Re: bar end PAS adjust knob:

How did you get it holding on handlbar end?!

I finally got this finished up this week and posted details in my build thread.

I am liking this hacked 'Auto Torq PAS' mode a lot!

 

[cloudcover]

hi there -

not 100% sure this is the right place to post this question -- happy to receive a re-direct if appropriate.

i've been using a throttle-controlled ebike for 2-3 years now and have been pretty happy with the experience for the most part. but since my goal is to use the motor just for assist, when needed, and because i live in an area with hills, the one part of the experience that i don't enjoy is having to regulate the throttle and having to hold the throttle down for extended bike rides. for that reason, i posted a question a year or two ago asking if there was a torque-sensing option that would work well for my needs.

i'm not sure the best way to describe what i'm looking for, but basically i want to be able to set a maximum level of torque that i am willing to apply through the pedals. below that maximum level, i'd like the motor to contribute basically nothing. but if the torque needed to get up a hill, for example, is greater than the maximum level of torque that i'm willing to apply, i'd like the controller to draw how much ever power is needed from the motor to make up for the excess requirement. so, for example, suppose i'm willing to expend X lb-ft of energy (or watts of energy or whatever is the best way to measure my output). and suppose that a flat road requires Y lb-ft of energy, where Y<X, then the motor would do nothing. but if a hilly road requires Z lb-ft of energy, where Z>X, then the motor would contribute Z-X of energy, and my pedaling would supply X energy, so the total would be the required Z amount of energy.

i don't follow ebike developments all the time, but noticed recently that ebikes.ca has a Thun bb sensor that can work with the Cycle Analyst V3. my question is whether those two things, in combination, can do what i have in mind? and if not, then as a practical matter, what's the experience like of using the thun sensor with the CA V3?

thanks!

 

[teklektik]

...the one part of the experience that i don't enjoy is having to regulate the throttle and having to hold the throttle down for extended bike rides
... basically i want to be able to set a maximum level of torque that i am willing to apply through the pedals.
... but if the torque needed to get up a hill, for example, is greater than the maximum level of torque that i'm willing to apply, i'd like the controller to draw how much ever power is needed from the motor to make up for the excess requirement.

There is no means for the bike to 'know' how much power is required to accomplish the task, whether you are assisting or not. How does it know if adding 5 Amps or 50Amps of assist is doing the job?

You can set the CA so that it will not assist until you are expending a minimum number of Human Watts, and you can set it so that it provides some amount of assistance proportional to your effort (100%, 150%, 300%, etc), but it can't just "make up for the excess requirement" because it doesn't know what that is - maybe it's a steep hill and you mentally decide the requirement is 20mph or maybe it's the flat and you decide the requirement is 10mph....

You can set the CA to maintain a fixed speed and then just pedal as much as wish to assist, but this can be a very disconcerting way to ride unless you have no terrain, pedestrian, or traffic considerations for long stretches of road. A similar effect can be had with the CA auto-cruise, but these modes can be like driving your car in traffic with cruise control engaged - there is a time and place for that feature to be beneficial.

Your best bet is use CA PAS with a Thun to simply multiply your expended Human Watts and then resort to the throttle when you need a little extra oomph to climb a hill. If your terrain is accommodating, you will be able to get by throttle-free much of the time. If necessary, you can add an 'Assist Level' knob so you can tweak the assist level to accommodate different situations over the course of your ride. In any case, you may find that eliminating just a part of your throttle usage relieves most the unpleasantness you now experience.

 

[izeman]

was this mentioned before? but i would like to request the cruise control to be a 'real' speed control but a throttle freeze.
eg. i ride 20km/h and release the throttle after the pre-defined seconds. depending on the throttle position it would become faster when i ride downhill and slower when going up.
but i would like to go 20km/h all the time.
is that possible?

 

[teklektik]

...i would like to request the cruise control to be a 'real' speed control
...i would like to go 20km/h all the time.
is that possible?

Yep - in the preset where you have cruise control enabled, configure ThrI->CntrlMode = Speed. This will make the throttle select a speed 0-100% of SLim->MaxSpeed , so set that parameter to something reasonable as well. Engaging auto-cruise will make the CA hold the speed set by the throttle - the CA will apply power as required to climb hills while maintaining that speed.

This works quite nicely, but driving with the throttle in Speed Throttle mode is a little strange. This might be best if you had a 'preset switch' so you could have a preset for 'normal throttle mode' with Current or Power Throttle, then quickly flip to the auto-cruise preset and Speed Throttle.

 

[hillyterrain]

Quick CA V3 throttle control question, I have the throttle control set to PasThrough but was hoping to use the ramp function, ramping the throttle up slowly, it is being ignored, is that because it is set to passthrough? I does currectly map the input 0.87 - 4.2 to a new output 1.35 to 3.2 but at a ramp rate of 0.05 V/Sec it should almost take a minute to get there, instead it is as instantaneous as having the throttle directly on the controller ...

any ideas? I'm on Beta 22, is that the problem?

 

[teklektik]

I have the throttle control set to PasThrough but was hoping to use the ramp function, ramping the throttle up slowly, it is being ignored,
... but at a ramp rate of 0.05 V/Sec it should almost take a minute to get there
... I'm on Beta 22, is that the problem?

Nope - it's not being ignored - the difficultly is that the ramping units are Sec/V not V/Sec so a setting of 0.05 is almost instantaneous.

Try something like 2.00 and things should work much better.

 

[hillyterrain]

Big thanks for pointing that out!!

 

[teklektik]

An updated version of the Unofficial User Guide is available for download.

The original plan was to update the doc after the official v3.0 product release, but PAS and other revisions have delayed things a bit. With the Spring build season upon us, there have been a flurry of downloads, but the posted doc had become pretty out of date. This intermediate version should get it caught up to the present v3.0 prelim5 firmware. A new update will be along as planned when the 3.0 firmware is finalized.

Enjoy!

 

[izeman]

teklektik: you really rock. Thanks for all the effort going into that document!

 

[Merlin]

another big thx from me

tekle, god of DS Gain

 

[hillyterrain]

yes, that document is amazing!

 

[teklektik]

Thanks for the kind words - but, in the end, the Guide is largely a compilation of posted material from this thread.

Many thanks to all who have posted questions, clever build or configuraton ideas, and especially updates on their experiences and problem resolutions.

 

[teklektik]

Oops - a couple of minor P5 changes were missing from the new Guide version and an anchored graphic got left behind when the text relocated to the next page. (I guess my proofreading was too much like speed reading... )

Fixed and re-posted. - Sorry 'bout that...